How to Connect Claude to Your Aforza Data
Ask three people in a commercial team where they check their numbers and you will get three different answers. The single front end is gone. What people expect now is to reach their data from wherever they happen to be working, and increasingly that place is an AI assistant.
The problem with data that only lives behind a screen
Most CPG commercial data sits behind a user interface. Getting an answer means knowing which dashboard, which report, which tab. So the question gets passed to someone else, and a five second answer becomes a two day round trip.
Teams that try to solve this by wiring AI tools into their systems usually hit the same wall twice. First, every tool needs its own integration, built and maintained separately. Second, and more seriously, they end up with a governance problem: which records can the AI see, acting as which user, under whose permissions? Most projects stall there, and the ones that do not stall often end up copying commercial data into a third place where nobody is quite sure who can read it.
What is the Aforza MCP?
The Aforza MCP is a Model Context Protocol connection to the Aforza platform. It allows an MCP compatible AI client, such as Claude, to read and act on Aforza data: accounts, order history, visits, promotions, pricing and more, using the same user permissions that apply inside Aforza itself.
MCP is an open standard, originally developed by Anthropic and now adopted across the AI industry, that gives AI applications a consistent way to connect to business systems. The Aforza MCP applies that standard to your commercial data, exposing Aforza’s objects and actions as tools an AI client can use. No middleware. No data copy. No separate AI vendor holding a duplicate of your commercial data.
How do you connect Claude to Aforza?
The connection is made through the Aforza MCP. The AI client authenticates against your Aforza environment, discovers the tools it is allowed to use, and from that point you simply ask questions in plain language. There is no configuration project, no pre built prompts, and no new interface for anyone to learn.
What can you actually do with it?
The demo below walks through three requests, made one after another with no setup in between.
An account overview
Ask for a summary of an account and its priorities for the quarter. Claude retrieves the account record, order history, visits and current promotional activity, then summarises where the focus should sit. The same account sits open in Aforza alongside it, showing the same numbers.
A daily visit plan
Ask for all visits planned today across the team. Claude returns the scheduled visits, the owners, the accounts and the objectives, matching the Aforza visit planner exactly.
A live promotion analysis
Ask for all promotions currently running at an account, ranked by sales revenue, and shown visually. Claude retrieves the active promotions and renders them as a ranked chart. Nothing was pre built. The view was generated in response to the question.
And it is not read only. The same connection can create and update records, so a request can end in an action rather than an answer.
Why the data model underneath matters more than the protocol
An MCP connection is only as useful as the data model it sits on top of. Ask a generic CRM for the promotions running at an account ranked by sales revenue and there is nothing meaningful to answer with, because promotions, mechanics, funds and depletions were never part of the model.
Aforza’s model is built for consumer goods from the ground up: trade promotions and trade spend, visit plans and call cycles, planogram compliance and share of shelf, order history, pricing conditions and route to market structures. That is what makes a question like “which promotions are underperforming at this account” answerable in one step rather than five.
It is also the same foundation that Aforza’s Vertical AI capability runs on. Inside the platform, purpose built CPG agents do the work where the work happens, prompting the rep in store, flagging distribution gaps, recommending where promotional investment should sit. Outside the platform, the MCP opens that same governed data and those same actions to whichever AI client a team already uses. The two are not alternatives. The agents make the platform intelligent; the MCP makes that intelligence reachable from anywhere.
Is the Aforza MCP secure?
Access runs through your existing Aforza security model and honours the same user profiles, permission sets, sharing rules and field level security you already have in place. A user cannot see anything through Claude that they could not see by logging into Aforza, and every request is subject to the same access controls and logging. Your data stays where it is.
Frequently asked questions
What is the Aforza MCP?
The Aforza MCP is a Model Context Protocol connection that lets AI clients such as Claude read and act on Aforza data, including accounts, orders, visits and promotions, using the user’s existing Aforza permissions.
Which AI clients can connect to Aforza through MCP?
Any MCP compatible client. Claude is the example shown here, and the same standard is supported by other AI applications and custom agents.
Does the Aforza MCP copy my data out of Salesforce?
No. Data is read live from your Aforza environment at the point of the request. Nothing is duplicated into a separate store.
Can the Aforza MCP write data, or only read it?
Both. It can retrieve records and it can take action, within the permissions of the user making the request.
Do I need to build anything to use it?
There is no configuration project or custom integration. The connection uses standard authentication and your existing Aforza access.
Getting started
If you are already an Aforza customer, the Aforza MCP works with your existing licences and access model. Speak to your account manager and we will walk you through connecting it and show you what your own data looks like on the other side.
Commercial Excellence Exchange
For more insights like this, join the Commercial Excellence Exchange, our community for leaders across Commercial Excellence, Sales Excellence, Field Effectiveness, Sales Effectiveness & Commercial Performance in the Consumer Goods industry.